A Preparatively Convenient Ligand-Free Catalytic PEG 2000 Suzuki-Miyaura Coupling Thomas M. Razler,*,† Yi Hsiao,† Feng Qian,‡ Ruiling Fu,‡ Rana Kashif Khan,†,§ and Wendel Doubleday† The Department of Process Research and DeVelopment and the Department of Biopharmaceutics Research and DeVelopment, Bristol-Myers Squibb, One Squibb DriVe, New Brunswick, New Jersey 08903
[email protected] ReceiVed October 10, 2008
A ligand-free Suzuki-Miyaura reaction for the crosscoupling of aryl and heteroaryl bromides with aryl and heteroarylboronic acids has been developed utilizing catalytic polyethylene glycol 2000 (PEG 2000). This preparatively convenient system afforded the corresponding cross-coupled products in good to excellent isolated yields after a simple aqueous workup. Transmission electron microscopy (TEM) analysis of the Pd-PEG 2000 catalyst system revealed in situgenerated palladium nanoparticles after only 1 min of reaction.
The palladium-catalyzed Suzuki-Miyaura cross-coupling reaction has emerged as an extremely powerful synthetic tool in a broad range of disciplines, ranging from the synthesis of novel materials to industrial manufacturing of pharmaceuticals.1,2 Significant research has been devoted toward the identification of environmentally friendly cross-coupling protocols. Specifically, low palladium loadings,3 phosphine-free catalyst systems,3-7 and microwave irradiation8 protocols in water have all been demonstrated to actively promote the Suzuki-Miyaura reaction. * To whom correspondence should be addressed. † Department of Process Research and Development. ‡ Department of Biopharmaceutics Research and Development. § Bristol-Myers Squibb 2007 Summer Intern. (1) (a) Miyaura, N. In Metal-Catalyzed Cross-Coupling Reactions; de Meijere, A., Diederich, F., Eds.; Wiley-VCH: New York, 2004; Chapter 2. (b) Miyaura, N.; Suzuki, A. Chem. ReV. 1995, 95, 2457. (c) Nakamura, I.; Yamamoto, Y. Chem. ReV. 2004, 104, 2127. (2) Rouhi, A. M. Chem. Eng. News 2004, 82, 49. (3) Liu, L.; Zhang, Y.; Xin, B. J. Org. Chem. 2006, 71, 3994. (4) Goodson, F. E.; Wallow, T. I.; Novak, B. M. Org. Synth. 1998, 75, 61. (5) Lu, F.; Ruiz, J.; Astruc, D. Tetrahedron Lett. 2004, 45, 9443. (6) Korolev, D. N.; Bumagin, N. A. Tetrahedron Lett. 2005, 46, 5751. (7) LeBlond, C. R.; Andrews, A. T.; Sun, Y.; Sowa, J. R., Jr. Org. Lett. 2001, 3, 1555.
10.1021/jo802277z CCC: $40.75 2009 American Chemical Society Published on Web 12/10/2008
Phase-transfer complex (PTC) mediated Suzuki-Miyaura cross-coupling reactions have recently become an area of interest.9 Used as solvents, the polyethylene glycol (PEG) PTCs promote ligand-free Suzuki-Miyaura reactions through reduction of Pd(II) to Pd(0),10,11 obviating the need for phosphine reducing agents.12 Preparative use of the current methods, however, is highly impractical due to the polar nature of PEG. The high viscosity of the reaction mixture renders reaction stirring and monitoring extremely difficult. More importantly, it is prohibitively difficult to isolate all but the most simple biaryl cross-coupled products from the PEG solvent by the standard diethyl ether extraction.10d,11a,b,12 Thus, a catalytic PEG crosscoupling protocol would minimize these difficulties, while retaining the beneficial metal stabilizing and solubilizing attributes of the PEGs. We report herein a practical ligand-free Suzuki-Miyaura cross-coupling protocol employing 1 mol % of Pd(OAc)2 and 10 mol % of PEG 2000. We began our investigation by screening the activity of a variety of PTCs under ligand-free reaction conditions (Table 1). To evaluate the relative activity of each PTC, a 1:1 mixture of 4-bromotoluene 1 and 4-methoxyphenylboronic acid 2 with 1 mol % of Pd(OAc)2 was reacted for 12 h. In the absence of a PTC, K2CO3 afforded 23% conversion to the cross-coupled product 3 (entry 1). In comparison to the other phase transfer complexes, the higher molecular weight PEG-based PTCs showed improved conversion. Specifically, PEG 2000-K2CO3 and PEG 2000-Cs2CO3 both delivered 42% conversion while PEG 12000-Cs2CO3 and PEG 20000-K2CO3 furnished 3 with 46% and 43% conversion, respectively. With these observations, we chose the PEG 2000-K2CO3 catalyst system due to its lower molecular weight, low cost, and ease of product isolation relative to the higher order PEG systems.13 In an attempt to boost reaction conversion, lithium halides14 (LiCl, LiBr, and LiI), solvent, and boronic acid stoichiometry were evaluated with the PEG 2000-K2CO3 catalyst system. LiBr afforded no change in conversion (ca. 42%) while LiCl and LiI delivered much lower yields of 3. Acetonitrile, methanol, and ethanol offered no advantage over tetrahydrofuran, however, a more concentrated THF and H2O reaction solution offered a substantial boost in reaction conversion (entries 15 and 16). Increasing the boronic acid stoichiometry to 1.3 equiv in conjunction with 1 mol % of Pd(OAc)2, 10 mol % of PEG 2000, 2 equiv of K2CO3, and a 1:1 ratio of THF and H2O at reflux delivered the highest conversion to cross-coupled product 3 (entry 16). Upon reaction completion, dilution with ethyl acetate (8) (a) Blettner, C. G.; Ko¨nig, W. A.; Stenzel, W.; Schotten, T. J. Org. Chem. 1999, 64, 3885. (b) Leadbeater, N. E.; Marco, M. Org. Lett. 2002, 4, 2973. (c) Leadbeater, N. E.; Marco, M. J. Org. Chem. 2003, 68, 888. (9) Lipshutz, B. H.; Petersen, T. B.; Abela, A. R. Org. Lett. 2008, 10, 1333. (10) (a) Li, J. H.; Hu, X. C.; Liang, Y.; Xie, Y. X. Tetrahedron 2006, 62, 31. (b) Corma, A.; Garcia´, A.; Leyva, A. J. Catal. 2006, 240, 87. (c) Li, J.-H.; Liu, W.-J.; Xie, Y.-X. J. Org. Chem. 2005, 70, 5409. (d) Liu, L.; Zhang, Y.; Wang, Y. J. Org. Chem. 2005, 70, 6122. (11) For the cross-coupling of aryl chlorides in PEG 400 solvent under oxygen or air, see: (a) Han, W.; Liu, C.; Jin, Z.-L. Org. Lett. 2007, 9, 4005. (b) Han, W.; Liu, C.; Jin, Z. AdV. Synth. Catal. 2008, 350, 501. (12) Luo, C.; Zhang, Y.; Wang, Y. J. Mol. Catal. A: Chem. 2005, 229, 7. (13) PEG 2000 powder is commercially available from Alfa Aesar at $78.00/ kg. (14) Lipshutz, B. H.; Sclafani, J. A.; Blomgren, P. A. Tetrahedron 2000, 56, 2139.
J. Org. Chem. 2009, 74, 1381–1384 1381
TABLE 1.
Evaluation of Phase Transfer Complexes (PTCs) in the Ligand-Free Suzuki-Miyaura Reactiona
entry 1 2b 3c 4d 5 6 7 8 9 10 11 15e 16 17 12 13 14 15f 16g
PTC
base
solvent
OABCDBr BCDBr ANCDCl Bu4NI 18-c-6 PEG 600 PEG 2000 PEG 2000 PEG 12000 PEG 20000 PEG 2000 PEG 2000 PEG 2000 PEG 2000 PEG 2000 PEG 2000 PEG 2000 PEG 2000
K2CO3 K2CO3 K2CO3 K2CO3 NaHCO3 K2CO3 Cs2CO3 K2CO3 Cs2CO3 Cs2CO3 K2CO3 K2CO3 K2CO3 K2CO3 K2CO3 K2CO3 K2CO3 K2CO3 K2CO3
THF/H2O THF/H2O THF/H2O THF/H2O THF/H2O THF/H2O THF/H2O THF/H2O THF/H2O THF/H2O THF/H2O ACN/H2O MeOH/H2O EtOH/H2O THF/H2O THF/H2O THF/H2O THF/H2O THF/H2O
additive
LiCl LiBr LiI
TABLE 2. Substrate Scope of Aryl and Heteroaryl Bromide 4 in the Reaction with 4-Methoxyphenylboronic Acid 2
conv, % 23 2.5 9.4 18 33 39 37 42 42 46 43 29 34 31 25 43 14 88 98
a Conditions: 4-bromotoluene 1 (1.0 mmol), 4-methoxyphenylboronic acid 2 (1.0 mmol), Pd(OAc)2 (0.01 mmol), PTC (0.1 mmol), base (2.0 mmol), solvent (5.0 mL), H2O (5.0 mL), reflux, 12 h. b O-Allyl-N-benzylcinchonidinium bromide (OABCDBr). c N-Benzylcinchonidinium bromide (BCDBr). d N-(9Anthracenylmethyl)cinchonidinium chloride (ANCDCl). e Acetonitrile (ACN). f 4-Bromotoluene (1.0 mmol), 4-methoxyphenylboronic acid (1.2 mmol), THF (1.6 mL), H2O (1.6 mL). g 4-Bromotoluene (1.0 mmol), 4-methoxyphenylboronic acid (1.3 mmol), THF (1.6 mL), H2O (1.6 mL).
followed by an aqueous brine wash enabled the efficient extraction of the product with no observable loss to the aqueous and PEG 2000 layers. Zhang10d and Liu11a,b have reported that oxygen can promote PEG-mediated Suzuki-Miyaura crosscoupling of simple aryl chlorides and bromides, respectively. In contrast, our reaction conducted open to the air resulted in the formation of homodimerization byproducts derived from the arylboronic acid in up to 12% yield. Presumably, air oxidation of Pd(0) to Pd(II) induces this dimerization pathway, which was not observed in the previous reports.15 An inert atmosphere of nitrogen completely suppressed the boronic acid dimerization and delivered the cross-coupled products in high yield. With conditions in hand, we explored the electronic and steric effects of various substituents in the cross-coupling of aryl and heteroaryl bromide coupling partner 4 with 4-methoxyphenylboronic acid 2 (Table 2). Aryl bromides with diverse electrondonating and electron-withdrawing substituents delivered the cross-coupled products in high yields (entries 1 to 9). Of note, 3-bromothioanisole (entry 6), a potentially difficult substrate due to palladium coordination, furnished the cross-coupled product 5e in 97% yield.16 The sterically demanding 2,6-tolylbromide (entry 3), however, proved to be a difficult substrate. After extended reaction times, product 5b was isolated in modest 22% yield with 61% and 53% recovery of the starting aryl bromide and boronic acid, respectively. (15) Miller, W. D.; Fray, A. H.; Quatroche, J. T.; Sturgill, C. D. Org. Proc. Res. DeV. 2007, 11, 359. (16) Chauhan, B. P. S.; Rathore, J. S.; Bandoo, T. J. Am. Chem. Soc. 2004, 126, 8493.
1382 J. Org. Chem. Vol. 74, No. 3, 2009
a General reaction conditions: aryl and heteroaryl bromide 4 (1.0 mmol), 4-methoxyphenylboronic acid 2 (1.3 mmol), Pd(OAc)2 (0.01 mmol), PEG 2000 powder (0.1 mmol), K2CO3 (2.0 mmol), THF (1.6 mL), H2O (1.6 mL), reflux. b Isolated yields. c 5-Bromoindole and 2 were recovered in 29% and 15% yield, respectively.
Heteroaryl halides have rarely been employed in ligand-free Suzuki-Miyaura cross-coupling protocols.17 Typically, high catalyst loading (5 mol %) is required to overcome catalyst deactivation through heteroatom coordination.18 With the 1 mol % of Pd(OAc)2 and catalytic PEG 2000 conditions, pyridine, pyrimidine, and pyrazole substrates all afforded their crosscoupled products in high isolated yields (entries 10 to 12). Modest conversion with 5-bromoindole, however, furnished 5l in 41% isolated yield, presumably due to palladium coordination by the indole nitrogen. In a similar fashion, the steric and electronic tolerance of the aryl and heteroarylboronic acid fragments were evaluated against 4-bromoanisole 6 (Table 3). Both electron-rich and electrondeficient arylboronic acids performed well under the reaction conditions (entries 1 to 7). Of note, Boc-protected 4-aminophenylboronic acid (entry 4) furnished 5p in 79% isolated yield as well as 15% yield of the des-Boc product. Presumably, desBoc 5p arises from basic hydrolysis of 5p, not from the starting boronic acid, as 4-aminophenylboronic acid was not observed (17) Conlon, D. A.; Pipik, B.; Ferdinand, S.; LeBlond, C. R.; Sowa, J. R., Jr.; Izzo, B.; Collins, P.; Ho, G.-J.; Williams, J. M.; Shi, Y.-J.; Sun, Y. AdV. Synth. Catal. 2003, 345, 931. (18) Thompson, A. E.; Hughes, G.; Batsanov, A. S.; Bryce, M. R.; Parry, P. R.; Tarbit, B. J. Org. Chem. 2005, 70, 388.
TABLE 3.
Substrate Scope of Aryl and Heteroarylboronic Acids 7 in the Reaction with 4-Bromoanisole 6
a General reaction conditions: 4-bromoanisole (1 mmol), boronic acid (1.3 mmol), Pd(OAc)2 (0.01 mmol), PEG 2000 powder (0.1 mmol), K2CO3 (2.0 mmol), THF (1.6 mL), H2O (1.6 mL), reflux. b Isolated yields. c 15% of the des-Boc coupled product was also isolated. d Des-Boc 5v was isolated in 22% yield. e Des-Boc 5x was isolated in 21% yield with 51% and 47% recovery of 6 and N-Boc-5-indoleboronic acid, respectively.
during the course of the reaction. Heteroarylboronic acids have had limited utility in the Suzuki-Miyaura coupling, primarily due to their difficult preparation, poor solubility, and low reactivity.19 Recently, the judicious choice of ligand, base, and solvent have been shown to have a dramatic effect on the reaction’s outcome.20 Evaluation of our Pd-catalytic PEG 2000 protocol against challenging heteroarylboronic acids revealed encouraging results. Specifically, coupling of 3,4-methylenedioxyphenylboronic acid (entry 8), benzofuran 2-boronic acid (entry 9), a substrate prone to proto-deborylation, and N-Bocpyrrole 2-boronic acid (entry 10) with 6 afforded the corresponding products 5t, 5u, and 5v in good yield. Cross-coupling of indoleboronic acid substrates, however, did not perform as well (entries 11 and 12). Multiple products were observed in the coupling of N-phenylsulfonyl-3-indoleboronic acid (entry 11), with no appreciable isolation of 5w. In the case of N-Boc(19) Tyrrell, E.; Brookes, P. Synthesis 2003, 4, 469. (20) (a) Kudo, N.; Perseghini, M.; Fu, G. C. Angew Chem., Int. Ed. 2006, 45, 1282. (b) Billingsley, K. L.; Anderson, K. W.; Buchwald, S. L. Angew. Chem., Int. Ed. 2006, 45, 3484. (c) Billingsley, K.; Buchwald, S. L. J. Am. Chem. Soc. 2007, 129, 3358.
5-indoleboronic acid (entry 12), only 21% of the des-Boc product was isolated with recovery of both the starting halide and boronic acid. Suzuki-Miyaura protocols employing a PEG solvent system have been shown to form palladium nanoparticles possessing high catalytic activity. 11a,b,12,21,22 Although such nanoparticles are normally prepared through the premixing of a palladium source and PEG at elevated temperatures, followed by the introduction of substrates and base after an induction period,10d,12,23 we set out to verify the involvement of palladium nanoparticles in our catalyst system. Indeed, examination of the catalytic PEG 2000 cross-coupling between 4-bromotoluene 1 and 4-methoxyphenylboronic acid 2 with transmission electron microscopy (TEM) revealed the presence of in situ-generated palladium nanoparticles (Figure 1). At 23 °C, 1 min after combining the reaction components, 10-20 nm palladium nanoparticles embedded upon the surface of a 200 nm PEG 2000 sphere were formed. At 5 min, the PEG 2000 sphere collapsed into well-dispersed 10-20 nm sized PdPEG 2000 nanoparticles with high surface area. At 30 min and 2 h, the nanoparticles began to cluster, forming agglomerates, with less surface area and presumably less catalytic activity.
FIGURE 1. TEM imaging of the Pd-PEG 2000 catalyst at (A) 1 min, (B) 5 min, (C) 30 min, (D) and 2 h.
To our knowledge, this represents the first observation of the changing nanoparticle agglomeration landscape over the course of a cross-coupling reaction. Interestingly, all reactions reported herein turned black within 1 min, indicating the precipitation of palladium from solution. Possibly, the larger chain length of PEG 2000 is the cause of the palladium black precipitation as the smaller PEG 400 has been reported to remain in solution with no observable palladium black formation for months.11b Currently, we are examining whether homogeneous or heterogeneous catalysis is operative with catalytic PEG 2000 and the effect the Pd-PEG 2000 nanoparticle agglomeration has on the reaction kinetics. In addition, evaluation of alternative crosscoupling partners such as aryl and heteroaryl chlorides and (21) For tetralkylammonium salt promoted Pd-nanoparticle formation see: Reetz, M. T.; Westermann, E. Angew Chem., Int. Ed. 2000, 39, 165. (22) Astruc, D.; Lu, F.; Aranzaes, J. R. Angew. Chem., Int. Ed. 2005, 44, 7852. (23) References 11a and 11b describe one of the first examples of in situgenerated nanoparticles.
J. Org. Chem. Vol. 74, No. 3, 2009 1383
iodides as well as potassium trifluoroborate salts and boronate esters is ongoing and will be reported in due course. In summary, we have developed a practical reaction protocol for the ligand-free, catalytic PEG 2000 Suzuki-Miyaura crosscoupling. Advantages of our new method include the facile isolation of polar cross-coupled products, low cost of PEG 2000, and the ability to promote the cross-coupling of heteroaryl bromides and heteroarylboronic acids. For the first time with catalytic PEG-based Suzuki-Miyaura reactions, we have observed the rapid formation (
